netdev - Re: [PATCH v1] rds: If one path needs re-connection, check all and re-connect

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <ba7da46b-a84d-142f-90e2-6b0be6899fbf@oracle.com>
Date:   Mon, 29 Jun 2020 10:55:28 -0700
From:   Rao Shoaib <rao.shoaib@...cle.com>
To:     David Miller <davem@...emloft.net>
Cc:     netdev@...r.kernel.org, linux-rdma@...r.kernel.org,
        ka-cheong.poon@...cle.com, david.edmondson@...cle.com
Subject: Re: [PATCH v1] rds: If one path needs re-connection, check all and
 re-connect


On 6/26/20 4:31 PM, David Miller wrote:
> From: rao.shoaib@...cle.com
> Date: Fri, 26 Jun 2020 11:34:38 -0700
>
>> +/* Check connectivity of all paths
>> + */
>> +void rds_check_all_paths(struct rds_connection *conn)
>> +{
>> +	int i = 0;
>> +
>> +	do {
>> +		rds_conn_path_connect_if_down(&conn->c_path[i]);
>> +	} while (++i < conn->c_npaths);
>> +}
> Please code this loop in a more canonial way:
>
> 	int i;
>
> 	for (i = 0; i < conn->c_npaths, i++)
> 		rds_conn_path_connect_if_down(&conn->c_path[i]);
>
> Thank you.

This was coded in this unusual way because the code is agnostic to the 
underlying transport. Unfortunately, IB transport does not 
initialize/use this field where as TCP does and counts starting from one.

If this is not acceptable, I would have to introduce a check for the 
transport or deal with zero count separately.

Let me know.

Regards,

Shoaib